Data exploration is a great place to start if you’re not familiar with the data set that you’re going to work with. You can use data exploration to create visualizations of your data that help better understand where relationships exists and you may want to move forward with predictive modeling.
Below are the prompts that Dacture will guide you through in the Data Exploration flow, along with additional information for each.
- Select a data set that you have uploaded, or upload a new data set.
- Choose Correlation Heat Map or Plot a Single Variable.
If you’ve chosen Correlation Heat Map:
- Select your data source from the drop-down menu.
- Click variables that you want to model or click Select All. Selected variables change color when you have clicked them. You can click a variable again to deselect it.
Choose a color palette from the drop-down menu. - Click Create. The heat map displays with a description to help with interpretation. You can use the About Your Data drop down to see additional information about the model and your data set.
- If you want to save the heat map, click Save Map and complete the Name and Description fields.
If you’ve chosen Plot a Single Variable:
- Select your data source from the drop-down menu.
- Click the variable you want to plot. The selected variable changes color. You can click on another variable to change which is selected, or click Clear to have none selected.
- Click Create. The plot of the variable appears below.
- If you want to save the plot, click Save Plot and complete the Name and Description fields.
- Optional: You can compare two single variable plots.
- Choose a plot from the Saved Plots dropdown.
- Click Compare Plots.